Other · model
Spark-X2.6-4B benchmark scores
As of 2026-10-29, Spark-X2.6-4B (Other) has measured scores on 21 benchmarks and estimated scores on 82 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 82.9% | estimated ± 2.7 pp, high confidence |
| AA-HLE | 15.0% | estimated ± 3.8 pp, high confidence |
| AA-SciCode | 37.2% | estimated ± 3.0 pp, high confidence |
| AIME26 | 91.2% | measured |
| CritPt | 0.4% | estimated ± 4.0 pp, high confidence |
| GDPval-AA | 19.5% | estimated ± 5.5 pp, medium confidence |
| GPQA Diamond | 77.1% | measured |
| HLE | 23.2% | estimated ± 6.6 pp, medium confidence |
| IFBench | 75.8% | measured |
| AA-LCR | 67.3% | measured |
| LiveCodeBench v6 | 87.4% | estimated ± 2.7 pp, high confidence |
| SciCode | 36.8% | measured |
| BrowseComp | 53.1% | measured |
| HLE w/ tools | 36.9% | measured |
| IFEval | 93.6% | measured |
| LongBench v2 | 53.3% | measured |
| VITA-Bench | 23.2% | estimated ± 6.7 pp, low confidence |
| τ²-bench results | 86.9% | measured |
| AA Coding Index | 30.4% | estimated ± 5.3 pp, medium confidence |
| APEX-Agents | 19.8% | estimated ± 0.6 pp, medium confidence |
| APEX-Agents-AA | 8.9% | estimated ± 4.2 pp, medium confidence |
| DeepSearchQA | 75.2% | estimated ± 7.3 pp, low confidence |
| SWE-bench Verified | 62.8% | measured |
| BFCL v4 | 71.8% | measured |
| CyberGym | 38.9% | estimated ± 7.3 pp, low confidence |
| JobBench | 34.4% | estimated ± 6.4 pp, medium confidence |
| SWE-bench Pro | 54.3% | measured |
| τ³-bench results | 34.7% | measured |
| WideResearch | 62.4% | estimated ± 3.5 pp, medium confidence |
| AA Agentic Index | 4.9% | estimated ± 5.9 pp, medium confidence |
| MMLU-Pro | 84.1% | estimated ± 2.5 pp, medium confidence |
| AA-IFBench | 74.9% | estimated ± 7.4 pp, medium confidence |
| Artificial Analysis Intelligence Index | 4.4% | estimated ± 6.1 pp, low confidence |
| GPQA | 69.9% | measured |
| MMLU | 87.9% | estimated ± 4.3 pp, high confidence |
| AA-Omniscience Accuracy | 21.7% | estimated ± 8.1 pp, medium confidence |
| MATH-500 | 97.1% | estimated ± 3.1 pp, medium confidence |
| MMMLU | 82.3% | estimated ± 1.1 pp, low confidence |
| AA MMLU-Pro | 84.8% | estimated ± 1.0 pp, low confidence |
| AIME 2025 | 91.9% | estimated ± 1.5 pp, medium confidence |
| Gert Labs | 33.5% | estimated ± 7.4 pp, medium confidence |
| AA Harvey LAB v1.0 | 81.5% | estimated ± 1.7 pp, low confidence |
| AA Tau3 Banking | 19.2% | estimated ± 3.9 pp, medium confidence |
| ARC-AGI-1 | 22.9% | estimated ± 4.0 pp, medium confidence |
| ARC-AGI-2 | 0.0% | estimated ± 11.6 pp, low confidence |
| CursorBench 3.1 | 50.8% | estimated ± 5.6 pp, low confidence |
| CursorBench 3.2 | 49.1% | estimated ± 3.6 pp, medium confidence |
| FrontierCode 1.1 Main | 37.0% | estimated ± 5.5 pp, medium confidence |
| FrontierMath v2 (Tier 4) | 3.8% | estimated ± 4.3 pp, low confidence |
| FrontierMath v2 (Tiers 1-3) | 16.9% | estimated ± 8.8 pp, low confidence |
| FrontierSWE v2 | 5.0% | estimated ± 9.0 pp, medium confidence |
| MazeBench | 0.0% | estimated ± 6.1 pp, low confidence |
| OSWorld-Verified | 72.7% | estimated ± 3.8 pp, medium confidence |
| Vals GPQA Diamond | 76.0% | estimated ± 2.0 pp, medium confidence |
| Vals LiveCodeBench | 83.1% | estimated ± 0.8 pp, low confidence |
| Vals MMLU-Pro | 79.1% | estimated ± 1.6 pp, medium confidence |
| Vals SWE-bench | 56.4% | estimated ± 5.3 pp, low confidence |
| AA AutomationBench | 28.6% | estimated ± 11.6 pp, low confidence |
| CWE-Bench-AA | 26.3% | estimated ± 5.2 pp, low confidence |
| AA Cyber Index | 26.0% | estimated ± 12.8 pp, low confidence |
| DeepsecBench-AA | 0.7% | estimated ± 7.4 pp, low confidence |
| AA ITBench | 32.9% | estimated ± 4.1 pp, medium confidence |
| CursorBench 4.0 | 31.1% | estimated ± 3.6 pp, medium confidence |
| DeepSWE | 15.1% | estimated ± 12.7 pp, low confidence |
| FrontierCode 2.0 | 18.8% | estimated ± 7.1 pp, low confidence |
| HLE w/o tools | 17.9% | measured |
| PostTrainBench v1.1 | 22.3% | estimated ± 4.7 pp, medium confidence |
| SWE Multilingual | 64.7% | measured |
| Toolathlon-Verified | 53.2% | estimated ± 5.7 pp, medium confidence |
| Vibe Code Bench | 0.6% | estimated ± 8.1 pp, low confidence |
| ARC-AGI-3 | 0.0% | estimated ± 10.6 pp, low confidence |
| C-Eval | 88.5% | estimated ± 1.2 pp, high confidence |
| Claw-Eval | 69.9% | measured |
| DeepPlanning | 41.3% | estimated ± 13.1 pp, low confidence |
| HMMT Feb 2025 | 91.7% | estimated ± 1.8 pp, low confidence |
| HMMT Feb 2026 | 81.3% | measured |
| HMMT Nov 2025 | 90.0% | estimated ± 2.8 pp, high confidence |
| INCLUDE | 82.2% | estimated ± 2.0 pp, low confidence |
| MCP Atlas | 62.4% | measured |
| MCP-Tasks | 76.6% | estimated ± 0.6 pp, low confidence |
| MMAnswerBench | 76.5% | estimated ± 1.3 pp, medium confidence |
| MMLU-ProX | 82.2% | estimated ± 1.3 pp, medium confidence |
| MMLU-Redux | 90.4% | estimated ± 1.7 pp, high confidence |
| NL2Repo | 40.0% | estimated ± 6.1 pp, medium confidence |
| QwenClawBench | 56.5% | estimated ± 2.6 pp, high confidence |
| SuperGPQA | 49.9% | estimated ± 8.2 pp, low confidence |
| Toolathlon | 26.8% | estimated ± 4.2 pp, medium confidence |
| LiveCodeBench Pro | 79.1% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (Text) | 30.6% | estimated ± 8.9 pp, low confidence |
| MMLU-Pro (Arcee) | 82.1% | estimated ± 3.4 pp, medium confidence |
| React Native Evals | 71.4% | estimated ± 2.3 pp, medium confidence |
| ResearchClawBench | 19.8% | estimated ± 1.3 pp, high confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, low confidence |
| DRACO | 69.1% | estimated ± 4.5 pp, low confidence |
| FrontierCode 1.1 Extended | 54.0% | estimated ± 3.0 pp, low confidence |
| HealthBench Professional | 26.1% | estimated ± 4.8 pp, low confidence |
| SimpleQA | 30.6% | estimated ± 14.6 pp, low confidence |
| LiveCodeBench | 68.0% | estimated ± 14.6 pp, low confidence |
| OpenHarmony Bench | 53.3% | estimated ± 4.9 pp, low confidence |
| IMOAnswerBench | 75.2% | measured |
| MRCRv2 | 59.9% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 60.8% | estimated ± 2.6 pp, low confidence |
| PinchBench | 88.1% | estimated ± 3.6 pp, medium confidence |